In Covid-19 Infection, Plasma Extracellular Vesicle Tissue Factor Activity Does Not Correlate with D-Dimer Levels

Objective: Identify a plasma-based activity, or biomarker, that defines the mechanism(s) by which Covid-19 disease triggers excessive coagulation. Introduction: While acute respiratory syndrome is the fundamental feature of severe Covid-19 disease, having a high level of the coagulation biomarker D-dimer upon admission is associated with increased thrombosis and mortality. As such, hospitalized patients are often placed on anticoagulant heparins. How Covid-19 triggers excessive coagulation is unresolved. Sars-CoV-2 infection could expose existing tissue factor (TF) to blood or, via cytokines, induce TF expression on cells that are in direct contact with blood. Extracellular vesicles (EV) are lipid bound microparticles released by all types of healthy and damaged cell and Covid-19 patient plasma EV TF activity has been recently reported. Cellular activation and damage due to SARS-CoV-2 could also release polyanionic nucleic acids and polyphosphates and generate neutrophil extracellular traps as contact surfaces for clot formation. Methods: Study 1. We attempted to identify excessive coagulation pathway activities in Covid-19 plasma-based, Ca++-induced thrombin generation assays. Assays were performed in the absence and presence of selective extrinsic (TF) and intrinsic (contact activation) pathway inhibitors (n=296 plasma samples). D-dimer levels were also determined. In a smaller study, Covid-19 patient samples were collected directly into citrate or citrate plus corn trypsin inhibitor, then processed for analysis. Study 2. We conducted studies to evaluate the extent to which EV TF activity contributes to the Covid-19-associated coagulopathies. Plasma EVs were isolated and EV TF activity determined by the difference in FXa activity in the absence vs presence of anti-TF antibody. D-dimer and tissue factor pathway inhibitor a (TFPIa) antigen levels were measured. Data from 232 samples collected from 96 Covid-19 positive patients and 18 samples from 14 healthy controls were analyzed. For each study analysis, patient samples were organized into groups based on the disease severity outcomes as follows: hospitalization (Hospitalization; n=37); intensive care (ICU; n=16); mechanical ventilation (Ventilation; n=22); or fatality (Deceased; n=22). Result: Study 1. Covid-19 samples showed considerable thrombin generation variability with some samples failing to generate thrombin; pathway selective inhibitors reduced thrombin generation while heparinase treatment increased thrombin generation. Upon analysis, thrombin generation parameters showed no significant correlations to either D-dimer levels or disease severity. Instead, plasma prepared from blood collected directly into corn trypsin inhibitor revealed that contact activation that occurred post-sample collection dominates procoagulant activity. Study 2. Figure 1, shows EV TF activities, D-dimer and TFPIα levels obtained for Covid-19 samples, with data segregated based on disease severity outcomes. Statistically significant difference versus the Hospitalized group are shown. TFPIa levels were highest in heparin IV patients (24.4+1.5 nM) vs Heparin-SQ (12.8+0.9 nM) vs enoxaparin (10.8 +0.7 nM) (p value:<0.0001). It is known that heparin treatment increases circulating TFPIα, however an increase in TFPIα might also further increase circulating TF/FVIIa/XaTFPI inhibitory complex, which would dissociate in citrated plasma, and might account for the increase in EV TF in other studies. Conclusions: Contact activation that occurs post-sample collection is sufficient to obscure endogenous triggers of coagulation, if present, in Covid-19 patients' plasma. D-dimer and TFPIα strongly correlate with disease severity although the latter is likely affected by heparin treatment. The most severe Covid-19 patients with high D-dimer did not show detectible plasma EV TF activity. Plasma EV TF activity does not appear to adequately represent the mechanism responsible for elevated D-dimer levels in Covid-19 cases. Figure 1 Figure 1. Disclosures Di Paola: CSL Behring: Consultancy, Honoraria.

Novel Thermostable Inhibitor of Contact Activation Tica Effectively Blocks Contact Activation in Low Tissue Factor Thrombin Generation

Background: Thrombin generation and other clotting assays suffer from a wide variation of pre-analytical variables. One of those pre-analytical variables is contact activation through blood withdrawal methods, different syringes, differences in blood coagulation tubes, blood transport and sample handling. It has been shown that the addition of contact activation inhibitors in low tissue factor activated thrombin generation leads to a correction of the, in these circumstances significant, increase in thrombin generation due to contact activation. We compare the novel 'thermostable inhibitor of contact activation' (TICA) to the current standard 'corn trypsin inhibitor' (CTI). Aim: Comparing the effectiveness of novel contact activation inhibitor TICA to the current standard CTI in low tissue factor-induced thrombin generation and recalcification in sodium citrate anticoagulated platelet poor plasma (PPP) and platelet rich plasma (PRP). Methods: We compared TICA, Corn trypsin inhibitor and plasma without contact activation inhibitors in low tissue factor PPP thrombin generation and in PRP recalcification thrombin generation, the latter the most sensitive condition for contact activation. In addition, we compared low tissue factor activated thrombin generation in plasma from severe hemophilia A patients with and without TICA during and after blood drawing. Thermostability - as a measure of shelf life - was measured and compared to CTI. Results: TICA is able to fully block contact activation in PRP recalcification experiments and is comparable to CTI in doing so. TICA significantly lowers low tissue factor induced thrombin generation by blocking contact activation. Pre-loading vacuum blood collection tubes with contact activation inhibitors is superior in inhibiting contact activation compared to addition of the inhibitor during the thrombin generation assay itself. TICA did not alter coagulation activity when added to FXIIa deficient plasma in thrombin generation. In contrast to CTI TICA is heat stable which will be of benefit to shelf life of pre-loaded blood drawing tubes. Conclusion: TICA is able to fully block contact activation and has several advantages over CTI. Disclosures No relevant conflicts of interest to declare.

Abnormal plasma clot formation and fibrinolysis reveal bleeding tendency in patients with partial factor XI deficiency

Key Points Platelet-poor plasma clotting and fibrinolysis assays detect bleeding tendency in patients with factor XI deficiency. Contact pathway inhibition with corn trypsin inhibitor increases sensitivity of these assays to bleeding tendency.

Neutrophil Extracelluar Traps Contributes to the Hypercoagulable State and Liver Damage in Liver Cirrhosis

Background: We recently reported that phosphatidylserine on blood and endothelial cells played an important role in the hypercoagulable state of liver cirrhosis (Wu et al, Liver Int 2016). Recent studies showed that enoxaparin could reduce the incidence of portal vein thrombosis and alleviate liver decompensation. Thus, we speculate that there may be other mechanisms involved in the hypercoagulability of cirrhosis. Neutrophil extracelluar traps (NETs) play an important role in thrombosis and organ dysfunction. Whether the benefit of enoxaparin is due to inhibiting NETs in cirrhotic patients remains unknown. Our objectives were to study the formation of NETs and their role in the hypercoagulable state and liver damage in cirrhotic patients. Methods:Cirrhotic patients (n = 36), healthy controls (n = 10), and mice treated with CCl4 (n = 50) were studied. Immunofluorescence confocal microscope was used to analyze and quantify NETs formation in neutrophil cells from human and mice. TAT and fibrin formation assays in normal and cirrhotic plasma were performed to analyze the procoagulant activity of NETs. Calibrated automated thrombography (CAT) was performed with and without 6 nM thrombomodulin I to analyze TM resistance changes induced by NETs. Liver injury was assessed by plasma alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activity, determined spectrophotometrically, and by morphometric evaluation of left lateral liver lobes. DNase I and activated protein C (APC) were used to abolish NETs. Results: We observed that cirrhotic platelets induced 1.9 fold higher NETs formation in neutrophil cells (p = 0.021). Increased NETs promoted higher TAT (Fig 1A, B) and fibrin formation in normal and cirrhotic plasma (p < 0.05). DNase I abolished the procoagulant activity of NETs in corn trypsin inhibitor (CTI ) or anti-TF pre-treated plasma (p < 0.001) (Fig 1C). TM pre-induced with NETs for 24 h showed a higher endogenous thrombin potential (ETP) than no treatment in normal plasma using CAT (p = 0.032). APC and de-N-sulfated heparin (de-N-Hep), but not DNase I, can decrease the ETP ratio (with/without TM) in cirrhotic plasma (p < 0.05) (Fig 1D). Mice treated with CCl4 showed a higher ALT, AST, necrotic area (Fig 2), and NETs formation than control (p < 0.05). DNase I and enoxaparin decreased ALT, AST, necrotic area (Fig 2) and NETs formation (p < 0.05). Conclusion s:Cirrhotic patients have enhanced NETs formation compared to healthy controls. NETs initiated intrinsic and extrinsic coagulation pathway and impaired anticoagulant function of TM. Additionally, NETs partly contributed to tissue injury in the liver of cirrhotic mice. DNase I and enoxaparin can disconnect NETs and decreased the toxin to liver. Figure 1 Neutrophil extracelluar traps (NETs) were isolated and incubated with plasma from healthy controls (A) or cirrhotic patents (B). TAT complexes were measured by ELISA. (C) TAT complexes formation of NETs were analyzed in normal plasma pre-treated with corn trypsin inhibitor (CTI ) or anti-TF in absence or presence of DNase I. (D) Calibrated automated thrombography was performed with and without 6 nM thrombomodulin (TM) in cirrhotic plasma. APC, activated protein C; de-N-Hep, de-N-sulfated heparin. *p < 0.05 vs. Control in figure A, B; *p < 0.001 vs. NETs in figure C; *p = 0.008 and #p = 0.039 vs. PBS in figure D. Figure 1. Neutrophil extracelluar traps (NETs) were isolated and incubated with plasma from healthy controls (A) or cirrhotic patents (B). TAT complexes were measured by ELISA. (C) TAT complexes formation of NETs were analyzed in normal plasma pre-treated with corn trypsin inhibitor (CTI ) or anti-TF in absence or presence of DNase I. (D) Calibrated automated thrombography was performed with and without 6 nM thrombomodulin (TM) in cirrhotic plasma. APC, activated protein C; de-N-Hep, de-N-sulfated heparin. *p < 0.05 vs. Control in figure A, B; *p < 0.001 vs. NETs in figure C; *p = 0.008 and #p = 0.039 vs. PBS in figure D. Figure 2 The necrotic area of liver tissue in control mice, cirrhotic mice treated with PBS, DNase I, enoxaparin or a combination of DNase I with enoxaparin. Enox, Enoxaparin. *p < 0.05 vs. PBS Figure 2. The necrotic area of liver tissue in control mice, cirrhotic mice treated with PBS, DNase I, enoxaparin or a combination of DNase I with enoxaparin. Enox, Enoxaparin. *p < 0.05 vs. PBS Disclosures No relevant conflicts of interest to declare.

Purified DNA, but Not Neutrophil Extracellular Traps (NETs), Promotes Contact Activation of Coagulation

Introduction Netosis refers to the extracellular release of nuclear material from neutrophils (PMNs) in the form of a meshwork called neutrophil extracellular traps (NETs). NETs are composed of a scaffold of DNA decorated with histones and specific cytoplasmic granules of neutrophils. Beside their ability to trap bacteria, NETs have been shown to promote thrombosis in animal models. Purified NET components have been suggested to enhance coagulation. However, the exact mechanisms by which intact NETs affect coagulation remain unclear. Objective To assess the effect of intact NETs on the activation of the contact system of coagulation. Methods Human PMNs were isolated from whole blood collected in 3.2% citrate (1 vol citrate / 9 vol blood) from normal volunteers using a negative selection by magnetic beads coated with antibodies directed against specific antigens of other blood cell types. Purity of PMN isolates was ≥ 96% while viability (assessed using trypan blue staining) was ≥ 98%. Isolated PMNs were re-suspended in RPMI + 0.5% bovine serum albumin or in normal pooled platelet-poor plasma (PPP) at 10-15,000 cells/µl final concentration. Re-suspended PMNs were stimulated to form NETs using phorbol myristate acetate (PMA: 600 nM) or ionophore (A23187: 5µM) and incubated at 37ᵒC in 5% CO2 for 3 hours. The effect of NETs was assessed on a synthetic construct of the contact system in-vitro using purified proteins (FXII, FXI, Prekallikrein and High Molecular Weight Kininogen) at physiologic concentrations in Hepes buffered saline containing 2mM Ca2+ and 0.1% polyethylene glycol (8000MW). The reaction was quenched with an equal volume of antithrombin (7uM), corn trypsin inhibitor (CTI, 0.2mg/mL) and heparin (1U/mL) at time points up to 1 hour after mixing with NETs. FXIa-antithrombin complexes were then detected using an in-house sandwich ELISA method. Thrombin generation (TG) was also monitored during 1 hour in re-calcified PPP containing stimulated PMNs with added 4µM synthetic phospholipid vesicles (PC:PE:PS). Finally, DNA was purified from neutrophils and its effect on contact activation was assessed in both the synthetic contact and plasma systems. Kaolin was used as positive control in both test systems; buffer without surface was used as the negative control in the synthetic contact system, and re-calcified PPP without added neutrophils was used as negative control in the TG assay. Results Massive amounts of NETs were released from stimulated PMNs as seen on fluorescence microscopy after staining for DNA using sytox green (Fig. I A&C). Extracellular release of NETs was confirmed by digestion following treatment with DNAse 1 (Fig. I B&D). Extracted DNA triggered contact activation (measured as XIa-antithrombin complexes) in buffer and enhanced TF-independent TG in PPP in a concentration dependent manner (Fig. I E&G). The latter was attenuated by corn trypsin inhibitor (0.1mg/mL) and abolished in FXII- or FXI-deficient plasma (Figure IE) indicating the dependency on the contact pathway activation. In contrast, intact NETs did not trigger contact activation in buffer and failed to enhance TG in re-calcified PPP in the absence of TF (Fig. I E&H). Conclusions Unlike purified DNA, intact NETs failed to activate coagulation via contact activation. These findings need to be confirmed using other readouts of contact system activation. These results do not exclude an effect of intact Nets on contact system amplification. Disclosures Monroe: Novo Nordisk: Honoraria, Research Funding.